| Summary: | 碩士 === 國立中山大學 === 企業管理學系研究所 === 97 === Initial public offerings (IPOs), especially common stock IPOs have drawn a lot of investors'' and researchers'' attentions for their short-run return rocketing phenomenon. Numerous articles focused on examining IPOs'' short- and long-run return structures in various methods and conclusions have been published. Ibbotson (1975), Ibbotson, Sindelar, and Ritter (1988) and Loughran, Ritter, and Rydqvist (1994) focused on examining whether IPOs did possess initial abnormal return, while Ritter (1991), Brav and Gompers (1997) and Eckbo and Norli (2005) contributed their efforts on explaining IPOs'' long-run return structures.
This thesis extended Eckbo and Norli''s (2005) study. I applied their model in examining Taiwan OTC IPOs'' long-run (5 years) return structures. The samples are dated from 1991 to 2002, a total of 261 IPOs (financial service companies excluded) are examined. I formed a portfolio which buys each IPO with offering prices in the first day of trading and sells them with closing prices on the trading day 5 years later. The equal-weighted returns are calculated and served as the daily raw return of the portfolio.
I used the Fama-French three factor model (size, book-to-market, RMRf) as the foundation, adding 2 factors (liquidity and leverage ratio) to the model and applying it to the samples. The outcomes are indicating that if the initial return was excluded and the portfolio return was calculated as the raw return minus risk-free return, the three-factor model displayed statistically significant factor loadings on size and RMRf factors while the intercept is significant as well. After adding liquidity and leverage ratio factors, all the factors in the model are significantly different from zero. The adjusted R-square values of the three- and five-factor models are 24.68% and 28.09%, respectively.
|
|---|
